Robust and safe control of mobile robots using control barrier functions: Experimental results

Ensuring accurate tracking for wheeled mobile robots (WMRs) while avoiding obstacles in complex environments is a crucial challenge. Existing methods often involve complex structures, suffer from high computational demands, or lack a unified mechanism to simultaneously ensure robust tracking, obstacle avoidance (OA), and safety under disturbances. Therefore, there is a need for a simple real-time control architecture that reduces computational burden while ensuring safe OA and robust trajectory tracking (TT). In response to this challenge, this paper proposes a novel robust safety tracking control approach to address the dual objectives of OA and TT under disturbances with two-fold ideas: 1) a robust nonsingular fast terminal sliding mode control (RNFTSMC) tracking algorithm integrated with an extended state observer (ESO), where the underactuated WMR model is first transformed into a linear canonical form using differential flatness, thereby facilitating the design of a stabilizing feedback controller to compensate for uncertainties and disturbances, ensure precise TT, minimize chattering, mitigate perturbations, avoid singularity issues, and achieve rapid convergence, and 2) a control barrier function (CBF)-based quadratic programming (QP) safety mechanism. The feedback controller is combined with safety constraints embedded in a QP formulation to ensure safe navigation in obstacle-rich environments and enable simultaneous robust TT and OA. The theoretical contribution lies in the formulation of robust tracking and safety controllers, along with stability and safety analysis proving finite-time convergence of tracking errors and forward invariance of the safe set under bounded disturbances. Finally, simulation and experimental results validate the proposed method’s effectiveness.